| { | |
| "best_metric": 0.45239612460136414, | |
| "best_model_checkpoint": "vit-base-food-items-v1/checkpoint-300", | |
| "epoch": 4.0, | |
| "eval_steps": 100, | |
| "global_step": 608, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06578947368421052, | |
| "grad_norm": 0.04839174449443817, | |
| "learning_rate": 0.00019671052631578949, | |
| "loss": 0.0259, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13157894736842105, | |
| "grad_norm": 5.086187362670898, | |
| "learning_rate": 0.00019342105263157894, | |
| "loss": 0.0743, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.19736842105263158, | |
| "grad_norm": 8.687716484069824, | |
| "learning_rate": 0.00019013157894736844, | |
| "loss": 0.0621, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2631578947368421, | |
| "grad_norm": 0.0554538369178772, | |
| "learning_rate": 0.00018684210526315792, | |
| "loss": 0.1584, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.32894736842105265, | |
| "grad_norm": 7.25691556930542, | |
| "learning_rate": 0.00018355263157894736, | |
| "loss": 0.0284, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.39473684210526316, | |
| "grad_norm": 0.0355791412293911, | |
| "learning_rate": 0.00018026315789473684, | |
| "loss": 0.1607, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.4605263157894737, | |
| "grad_norm": 6.474045276641846, | |
| "learning_rate": 0.00017697368421052632, | |
| "loss": 0.2034, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.5263157894736842, | |
| "grad_norm": 0.047177255153656006, | |
| "learning_rate": 0.0001736842105263158, | |
| "loss": 0.1755, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5921052631578947, | |
| "grad_norm": 7.999953269958496, | |
| "learning_rate": 0.00017039473684210527, | |
| "loss": 0.078, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6578947368421053, | |
| "grad_norm": 0.2906012237071991, | |
| "learning_rate": 0.00016710526315789475, | |
| "loss": 0.1773, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6578947368421053, | |
| "eval_accuracy": 0.8472727272727273, | |
| "eval_loss": 0.7279737591743469, | |
| "eval_runtime": 6.7097, | |
| "eval_samples_per_second": 81.971, | |
| "eval_steps_per_second": 10.284, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.7236842105263158, | |
| "grad_norm": 0.038031741976737976, | |
| "learning_rate": 0.00016381578947368422, | |
| "loss": 0.1011, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.7894736842105263, | |
| "grad_norm": 0.8751915097236633, | |
| "learning_rate": 0.0001605263157894737, | |
| "loss": 0.1059, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.8552631578947368, | |
| "grad_norm": 0.08943302929401398, | |
| "learning_rate": 0.00015723684210526318, | |
| "loss": 0.0334, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.9210526315789473, | |
| "grad_norm": 0.17175784707069397, | |
| "learning_rate": 0.00015394736842105265, | |
| "loss": 0.1515, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.9868421052631579, | |
| "grad_norm": 0.053591687232255936, | |
| "learning_rate": 0.0001506578947368421, | |
| "loss": 0.1301, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0526315789473684, | |
| "grad_norm": 0.026137366890907288, | |
| "learning_rate": 0.00014736842105263158, | |
| "loss": 0.0102, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.118421052631579, | |
| "grad_norm": 0.09105370193719864, | |
| "learning_rate": 0.00014407894736842106, | |
| "loss": 0.0066, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.1842105263157894, | |
| "grad_norm": 0.050408605486154556, | |
| "learning_rate": 0.00014078947368421053, | |
| "loss": 0.0679, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.051493316888809204, | |
| "learning_rate": 0.0001375, | |
| "loss": 0.007, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.3157894736842106, | |
| "grad_norm": 0.023582015186548233, | |
| "learning_rate": 0.00013421052631578948, | |
| "loss": 0.0589, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.3157894736842106, | |
| "eval_accuracy": 0.8872727272727273, | |
| "eval_loss": 0.5529205203056335, | |
| "eval_runtime": 5.9487, | |
| "eval_samples_per_second": 92.458, | |
| "eval_steps_per_second": 11.599, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.381578947368421, | |
| "grad_norm": 0.0221235528588295, | |
| "learning_rate": 0.00013092105263157893, | |
| "loss": 0.0046, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.4473684210526316, | |
| "grad_norm": 6.497156620025635, | |
| "learning_rate": 0.00012763157894736844, | |
| "loss": 0.0086, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.513157894736842, | |
| "grad_norm": 0.013416736386716366, | |
| "learning_rate": 0.00012434210526315791, | |
| "loss": 0.0042, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.5789473684210527, | |
| "grad_norm": 0.012088390998542309, | |
| "learning_rate": 0.00012105263157894738, | |
| "loss": 0.1094, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.6447368421052633, | |
| "grad_norm": 7.198599338531494, | |
| "learning_rate": 0.00011776315789473684, | |
| "loss": 0.045, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.7105263157894737, | |
| "grad_norm": 0.031135905534029007, | |
| "learning_rate": 0.00011447368421052632, | |
| "loss": 0.0331, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.776315789473684, | |
| "grad_norm": 0.07299932837486267, | |
| "learning_rate": 0.0001111842105263158, | |
| "loss": 0.0535, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.8421052631578947, | |
| "grad_norm": 0.02018345519900322, | |
| "learning_rate": 0.00010789473684210527, | |
| "loss": 0.0241, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.9078947368421053, | |
| "grad_norm": 0.012926718220114708, | |
| "learning_rate": 0.00010460526315789475, | |
| "loss": 0.033, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.973684210526316, | |
| "grad_norm": 8.804197311401367, | |
| "learning_rate": 0.00010131578947368421, | |
| "loss": 0.043, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.973684210526316, | |
| "eval_accuracy": 0.9090909090909091, | |
| "eval_loss": 0.45239612460136414, | |
| "eval_runtime": 5.7174, | |
| "eval_samples_per_second": 96.197, | |
| "eval_steps_per_second": 12.068, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.039473684210526, | |
| "grad_norm": 21.062307357788086, | |
| "learning_rate": 9.802631578947369e-05, | |
| "loss": 0.0327, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.1052631578947367, | |
| "grad_norm": 0.009257642552256584, | |
| "learning_rate": 9.473684210526316e-05, | |
| "loss": 0.0151, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.1710526315789473, | |
| "grad_norm": 0.014151917770504951, | |
| "learning_rate": 9.144736842105264e-05, | |
| "loss": 0.0248, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.236842105263158, | |
| "grad_norm": 0.013802828267216682, | |
| "learning_rate": 8.81578947368421e-05, | |
| "loss": 0.003, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.3026315789473686, | |
| "grad_norm": 0.014456182718276978, | |
| "learning_rate": 8.486842105263159e-05, | |
| "loss": 0.0035, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.3684210526315788, | |
| "grad_norm": 0.006758903618901968, | |
| "learning_rate": 8.157894736842105e-05, | |
| "loss": 0.0024, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.4342105263157894, | |
| "grad_norm": 0.009314753115177155, | |
| "learning_rate": 7.828947368421053e-05, | |
| "loss": 0.0024, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.006471664644777775, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0022, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.5657894736842106, | |
| "grad_norm": 0.013896413147449493, | |
| "learning_rate": 7.171052631578947e-05, | |
| "loss": 0.0023, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.6315789473684212, | |
| "grad_norm": 0.009549788199365139, | |
| "learning_rate": 6.842105263157895e-05, | |
| "loss": 0.0022, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.6315789473684212, | |
| "eval_accuracy": 0.8909090909090909, | |
| "eval_loss": 0.5150398015975952, | |
| "eval_runtime": 6.2356, | |
| "eval_samples_per_second": 88.203, | |
| "eval_steps_per_second": 11.065, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.6973684210526314, | |
| "grad_norm": 0.00833881739526987, | |
| "learning_rate": 6.513157894736842e-05, | |
| "loss": 0.0024, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.763157894736842, | |
| "grad_norm": 0.006957135163247585, | |
| "learning_rate": 6.18421052631579e-05, | |
| "loss": 0.0021, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.8289473684210527, | |
| "grad_norm": 0.006556599400937557, | |
| "learning_rate": 5.855263157894737e-05, | |
| "loss": 0.0021, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.8947368421052633, | |
| "grad_norm": 0.007122657261788845, | |
| "learning_rate": 5.526315789473685e-05, | |
| "loss": 0.0021, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.9605263157894735, | |
| "grad_norm": 0.0069893728941679, | |
| "learning_rate": 5.197368421052632e-05, | |
| "loss": 0.002, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.026315789473684, | |
| "grad_norm": 0.006159682292491198, | |
| "learning_rate": 4.868421052631579e-05, | |
| "loss": 0.002, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.0921052631578947, | |
| "grad_norm": 0.0069947754964232445, | |
| "learning_rate": 4.539473684210527e-05, | |
| "loss": 0.0019, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.1578947368421053, | |
| "grad_norm": 0.007419601548463106, | |
| "learning_rate": 4.210526315789474e-05, | |
| "loss": 0.0018, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.223684210526316, | |
| "grad_norm": 0.006330096162855625, | |
| "learning_rate": 3.8815789473684214e-05, | |
| "loss": 0.0018, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.2894736842105265, | |
| "grad_norm": 0.006105512380599976, | |
| "learning_rate": 3.5526315789473684e-05, | |
| "loss": 0.0018, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.2894736842105265, | |
| "eval_accuracy": 0.9018181818181819, | |
| "eval_loss": 0.49247637391090393, | |
| "eval_runtime": 6.5136, | |
| "eval_samples_per_second": 84.439, | |
| "eval_steps_per_second": 10.593, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.3552631578947367, | |
| "grad_norm": 0.006337973289191723, | |
| "learning_rate": 3.223684210526316e-05, | |
| "loss": 0.0018, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.4210526315789473, | |
| "grad_norm": 0.005863433238118887, | |
| "learning_rate": 2.8947368421052634e-05, | |
| "loss": 0.0018, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.486842105263158, | |
| "grad_norm": 0.0057103936560451984, | |
| "learning_rate": 2.565789473684211e-05, | |
| "loss": 0.0017, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.5526315789473686, | |
| "grad_norm": 0.004713858477771282, | |
| "learning_rate": 2.236842105263158e-05, | |
| "loss": 0.0018, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.6184210526315788, | |
| "grad_norm": 0.007430619560182095, | |
| "learning_rate": 1.9078947368421056e-05, | |
| "loss": 0.0017, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.6842105263157894, | |
| "grad_norm": 0.0051925876177847385, | |
| "learning_rate": 1.5789473684210526e-05, | |
| "loss": 0.0018, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 0.0064788335002958775, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.0017, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.8157894736842106, | |
| "grad_norm": 0.006365407258272171, | |
| "learning_rate": 9.210526315789474e-06, | |
| "loss": 0.0017, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 3.8815789473684212, | |
| "grad_norm": 0.005164624657481909, | |
| "learning_rate": 5.921052631578948e-06, | |
| "loss": 0.0018, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.9473684210526314, | |
| "grad_norm": 0.006292811129242182, | |
| "learning_rate": 2.631578947368421e-06, | |
| "loss": 0.0017, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.9473684210526314, | |
| "eval_accuracy": 0.9018181818181819, | |
| "eval_loss": 0.4941176474094391, | |
| "eval_runtime": 6.4553, | |
| "eval_samples_per_second": 85.201, | |
| "eval_steps_per_second": 10.689, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 608, | |
| "total_flos": 7.501829674622976e+17, | |
| "train_loss": 0.03790271527280933, | |
| "train_runtime": 250.8529, | |
| "train_samples_per_second": 38.588, | |
| "train_steps_per_second": 2.424 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 608, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.501829674622976e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |